Osteoarthritis (OA) is a highly prevalent chronic disease leading to significant functional limitations in both males and females, but particularly women. We propose to characterize the natural history of osteoarthritis (of the knee and hand) using radiographs, interviews and markers of cartilage and bone turnover as well as joint inflammation with this longitudinal study. The specific questions are: 1.Do biochemical markers of osteoarthritis (OA) provide evidence os OA earlier than radiographs.? 2.Can turnover markers be used to define natural history and progression of arthritis? 3.Are bone mineral density(BMD) loss and development/initiation of OA highly regulated? These questions can be addressed efficiently by concatenating historical data from two previously generated population-based groups. One population(Tecumseh Bone Health Study) of 573 women was 25-45 years at their 1992 baseline evaluation (R01-AR-40888--Bone Mineral Density Change and the Climacteric). Hand and knee films were taken two times four years apart (1992 and 1996) along with an annual BMD measurement. Annual urine and serum specimens were collected and are available for analysis of OA markers. The second group, from the SWAN Study (NR-04061), is a population-based group of 300 African-American and 150 Caucasian pre and perimenopausal women, aged 42-52 years at their 1996 baseline when hand and knee films were characterized and serum and urine collected. To the retrospectively available data, we propose to recontract these 1,023 women for radiographs (hand and knee) and interviews in 1998 and 2000 and add annual blood and urine collection with identification of potential markers of arthritis (including turnover on bone/collagen and inflammation). This would allow the examination of the initiation of osteoarthritis using radiographs, interviews and turnover biomarkers. This information about the natural history of osteoarthritis should allow us to consider more appropriate prevention and intervention strategies and offer the potential to identify markers prognostic of disease incidence and of processes involved in its pathobiology.